Low-concentration Homogenized Castor Oil
`Eye Drops for Noninflamed Obstructive
`Meibomian Gland Dysfunction
`
`Eiki Goto, MD,1,2 Jun Shimazaki, MD,1 Yu Monden, MD,3 Yoji Takano, MD,1,2 Yukiko Yagi,1
`Shigeto Shimmura, MD,1,2 Kazuo Tsubota, MD1,2
`
`Objective: We developed low-concentration homogenized castor oil eye drops for the treatment of patients
`with noninflamed obstructive meibomian gland dysfunction (MGD), a major cause of lipid-deficiency dry eye, and
`assessed the safety, stability, and efficacy of the eye drops.
`Design: Randomized, double-masked, placebo-controlled crossover clinical trial.
`Participants: Forty eyes of 20 patients with noninflamed MGD.
`Methods: After a preliminary study of eye drops containing castor oil, 2% castor oil and 5% polyoxyethylene
`castor oil (emulsifier) were mixed to formulate homogenized oil eye drops. The patients were assigned randomly
`to receive oil eye drops or placebo six times daily for 2 periods of 2 weeks each.
`Main Outcome Measures: At the end of each treatment period, we assessed symptoms, tear interference
`grade, tear evaporation, fluorescein and rose bengal scores, tear break-up time (BUT), and meibomian gland
`orifice obstruction. Safety and stability tests were also performed.
`Results: Symptom scores, tear interference grade, tear evaporation test results, rose bengal scores, tear
`BUT, and orifice obstruction scores after the oil eye drop period showed significant improvement compared with
`the results after the placebo period. No complications attributable to the eye drops were observed. The oil eye
`drops were stable when stored at 4°C.
`Conclusions: The results indicate that castor oil eye drops are effective and safe in the treatment of MGD.
`The possible mechanisms of this treatment are improvement of tear stability as a result of lipid spreading, ease
`of meibum expression, prevention of
`tear evaporation, and the lubricating effect of
`the oil eye
`drops. Ophthalmology 2002;109:2030 –2035 © 2002 by the American Academy of Ophthalmology, Inc.
`
`Meibomian glands secrete lipids into tears, and the lipids
`expand to form the oily preocular tear film layer that is
`responsible for preventing excessive evaporation of tears
`while also providing a barrier function at the lid margin,
`lubrication during blinking, and producing a smooth optical
`surface.1–5 Noninflamed obstructive meibomian gland dys-
`function (MGD) is the major cause of lipid tear deficiency
`or evaporative dry eye and recently has attracted attention as
`
`Originally received: October 29, 2001.
`Manuscript no. 210908.
`Accepted: March 22, 2002.
`1 Department of Ophthalmology, Tokyo Dental College, Chiba, Japan.
`2 Department of Ophthalmology, Keio University School of Medicine,
`Tokyo, Japan.
`3 Department of Ophthalmology, Kurume University School of Medicine,
`Kurume, Japan.
`Supported by grants from the Japanese Ministry of Education and Science,
`Tokyo, Japan; Medical School Faculty and Alumni Grants of Keio Uni-
`versity Medical Science Fund, Tokyo, Japan; Hightech Research Center at
`Tokyo Dental College, Chiba, Japan; and Nihon Tenganyaku Kenkyusho
`Co. Ltd., Nagoya, Japan.
`Authors Eiki Goto, MD, Kazuo Tsubota, MD, and Nihon Tenganyaku
`Kenkyusho Co., Ltd., are in the process of obtaining a Japanese patent on
`the eye drops described herein and their clinical application.
`Correspondence and reprint requests to Kazuo Tsubota, MD, Department
`of Ophthalmology, Tokyo Dental College, 5-11-13 Sugano, Ichikawa-shi,
`Chiba, Japan 272-8513. E-mail: eikigoto@nifty.com.
`
`2030
`
`© 2002 by the American Academy of Ophthalmology, Inc.
`Published by Elsevier Science Inc.
`
`a cause of ocular discomfort.6 – 8 Meibomian gland dysfunc-
`tion causes a decreased lipid supply, which in turn leads to
`increased tear evaporation, decreased tear stability, loss of
`lubrication, and damage to ocular surface epithelium, result-
`ing in symptoms related to dry eye.7–9 The conventional
`treatment for blepharitis and MGD includes warm com-
`presses,
`lid hygiene, and topical or systemic medica-
`tion.5,10 –12 Nevertheless, long-term compliance with con-
`ventional therapy is often insufficient and thus results in
`only limited relief of symptoms. Another possible approach
`is direct supplementation with oil as the deficient compo-
`nent. Lipid-containing eye drops designed to simulate nat-
`ural tears previously have been reported to be an effective
`treatment for dry eye, but they have not been accepted
`generally.13,14 Because high-concentration oil eye drops and
`ointments are viscous and cause blurred vision, resulting in
`decreased patient satisfaction, we conducted a preliminary
`assessment of the efficacy of a low-concentration castor oil
`mixture, and in this study assessed the efficacy, safety, and
`stability of low-concentration and homogenized oil eye drops.
`
`Materials and Methods
`
`Preliminary Study to Choose Oil
`We used seven oil ingredients permitted as additives for eye drops
`and ointments in Japan to formulate a low-concentration mixture
`
`ISSN 0161-6420/02/$–see front matter
`PII S0161-6420(02)01262-9
`
` EXHIBIT 1017
`
`

`
`Goto et al
`
`䡠 Castor Oil Eye Drops for MGD
`
`Table 1. Results of Preliminary Study using Castor Oil Mixture
`for Noninflamed Obstructive Meibomian Gland Dysfunction
`(Mean ⫾ Standard Deviation)
`
`Examination
`
`Face score (1–9)
`Tear evaporation rates
`(⫻10⫺7 g/second)
`Fluorescein score (0–9)
`Rose bengal score (0–9)
`Tear break-up time
`(sec)
`Schirmer’s test value (mm)
`
`Before*
`7.5 ⫾ 1.6
`30 ⫾ 9.0
`
`1.6 ⫾ 1.5
`1.2 ⫾ 1.4
`3.4 ⫾ 2.4
`
`After†
`5.3 ⫾ 1.4‡
`22 ⫾ 11‡
`
`0.20 ⫾ 0.60‡
`0.50 ⫾ 0.76
`9.0 ⫾ 3.6‡
`
`P
`Value
`
`0.01
`0.02
`
`0.005
`0.2
`0.002
`
`18 ⫾ 14
`
`16 ⫾ 12
`
`0.6
`
`Data were analyzed by Wilcoxon’s signed rank test for nonparametric
`paired data. A P value of ⬍0.05 was accepted as statistically significant.
`* Result before use of castor oil mixture.
`† Result after 2 weeks of application of castor oil mixture.
`‡ Statistically significant.
`
`with preservative-free artificial tears (Soft Santear; Santen Phar-
`maceutical Co, Ltd., Osaka, Japan) and tested them on normal and
`dry eye volunteers.15 Because the subjects reported that the castor
`oil (Astra Japan, Osaka, Japan) mixture eye drops were the most
`comfortable, this mixture was formulated for use in a preliminary
`trial and applied to 8 MGD patients 6 times daily for 2 weeks. The
`ocular surface and tear function were examined and compared
`before and after application. The examinations are described in
`detail in a later section. As shown in Table 1, the face score,
`fluorescein score, tear break-up time (BUT), and tear evaporation
`rates improved significantly. The mixture was simple to formulate,
`but being made from only hydrophobic oil, it may not spread
`readily over the ocular surface. We therefore decided to formulate
`homogenized oil eye drops with castor oil.
`
`Formulation, Safety, and Stability of Low-
`concentration Homogenized Oil Eye Drops
`To formulate low-concentration and homogenized oil eye drops
`(oil eye drops), 2% castor oil lubricant (castor oil; Yoshida Phar-
`maceutical Co., Tokyo, Japan), 5% polyoxyethylene castor oil
`(POE castor oil [emulsifier]; Sigma, St. Louis, MO), 0.3% sodium
`chloride, 0.15% potassium chloride, and 0.5% boric acid were
`stirred into distilled water under sterile conditions. The concentra-
`tion of castor oil in the preliminary study was changed to formulate
`an emulsion. The characteristics of the eye drops were: lucent,
`odorless, and low viscosity. The pH of the eye drops was 7.0, and
`their osmolarity was 260 mOsm/kg H2O. Bacterial and fungal
`cultures of the remaining unused eye drops were performed at the
`end of therapy under the usual conditions at room temperature.16
`The stability of the eye drops was assessed by testing for 4 weeks
`at 60°C and 4°C. Stress testing (testing at 60°C for 4 weeks) was
`performed instead of accelerated testing (testing at 40°C for 6
`months) for the eye drops.17 In these tests, the pH of the eye drops
`was measured, and the volunteers were checked for any irritation
`that might have been caused by administering the eye drops. The
`samples stored at 4°C remained stable at pH 7.0 for 4 weeks and
`did not cause irritation, whereas the pH value of the samples stored
`at 60°C changed to 5.8 and caused irritation.
`
`Subjects
`We examined a consecutive series of 20 patients (40 eyes) with
`MGD whose symptoms had not improved sufficiently despite
`
`conventional treatment, such as by lid hygiene and topical therapy
`with artificial tears, antibiotics, and corticosteroids or systemic
`antibiotics. Both eyes of patients were included. The participants
`consisted of 7 males and 13 females (52.1 ⫾ 11.0 years) with
`MGD. Nine patients had a Schirmer’s test value of less than 5 mm,
`indicating aqueous tear deficiency.9,18 Four patients of the aqueous
`tear deficiency group were diagnosed with Sjo¨gren’s syndrome
`according to aqueous tear deficiency.19 Eyes with anterior bleph-
`aritis of more than moderate severity, infectious conjunctivitis,
`MGD with acute inflammation, and eyes with excessive expression
`of meibum (seborrheic MGD) were excluded from the study. No
`patients used contact lenses.
`
`Criteria for Meibomian Gland Dysfunction
`Criteria for the diagnosis of MGD were: presence of meibomian
`gland dropout, poor meibum expression, and lack of active inflam-
`mation.7–9 Transillumination examination (meibography) with a
`fiber-optic device (L-3920; Inami Co., Tokyo, Japan) was per-
`formed.20 Visible loss of meibomian gland structure (gland drop-
`out) was observed during meibography and was considered evi-
`dence of MGD, because this finding was reported to be a good
`parameter for MGD-associated ocular surface changes.7–9 The
`degree of meibomian gland dropout was scored as described
`previously7-9: grade 0, no gland dropout; grade 1, gland dropout in
`less than half of the inferior tarsus; and grade 2, gland dropout in
`more than half of the inferior tarsus.
`To assess meibum expression and meibomian gland orifice
`obstruction, digital pressure was applied on the upper tarsus, and
`the degree of ease of expression of meibomian secretions
`(meibum) was evaluated semiquantitatively as follows: grade 0,
`clear meibum easily expressed; grade 1, cloudy meibum expressed
`with mild pressure; grade 2, cloudy meibum expressed with more
`than moderate pressure; and grade 3, meibum could not be ex-
`pressed even with strong pressure.7,9 Patients with a meibography
`score of grade 1 or 2 and meibomian gland orifice obstruction
`score of 2 or 3 were diagnosed as having MGD in this study.
`
`Assessment of Symptoms, Tears, Ocular Surface,
`and Meibomian Glands
`A questionnaire on overall comfort, complaints, and complications
`was administered by using a face score card that showed nine
`faces, each with a different expression.16 For example, the saddest
`face (scored as 9) described severe symptoms and irritation of the
`ocular surface, and the happiest face (scored as 1) represented no
`irritation of the ocular surface. This face score was used to assess
`patient sensation of lubrication and smoothness during blinking in
`this study.
`Tear interference images were assessed using a DR-1 camera
`(Kowa, Ltd., Nagoya, Japan) at ⫻12 magnification based on the
`grading previously reported: grade 1, somewhat gray color, uni-
`form distribution; grade 2, somewhat gray color, nonuniform dis-
`tribution; grade 3, a few colors, nonuniform distribution; grade 4,
`many colors, nonuniform distribution; grade 5, corneal surface
`partially exposed.21,22 Tear evaporation during normal blinking
`was measured by the method previously reported.23 The evapora-
`tion rate at 40% ambient humidity was used as a representative
`value.
`The ocular surface was examined by the double staining
`method. A 2-␮l volume of preservative-free solution consisting of
`1% fluorescein and 1% rose bengal dye was applied to the con-
`junctival sac. The intensity of the rose bengal staining of the
`cornea and conjunctiva was recorded, with a maximum score of 9.
`Fluorescein staining of the cornea was also rated from 0 to 9. Tear
`BUT was measured three times, and the measurements were av-
`
`2031
`
`

`
`Ophthalmology Volume 109, Number 11, November 2002
`
`Table 2. Schedule of Examinations in Crossover Study of Castor Oil Eye Drops for Noninflamed Obstructive Meibomian Gland
`Dysfunction
`
`Group 1
`Group 2
`
`0
`Wash-out Period Began
`AT
`AT
`
`Week
`
`2
`Beginning of Study
`Oil eye drop* usage began
`Placebo† usage began
`
`4
`Patients Switched Eye Drops
`Switched to placebo†
`Switched to oil eye drops*
`
`6
`Completion of Study
`Completed
`Completed
`
`AT ⫽ preservative-free artificial tear.
`* Low-concentration homogenized castor oil eye drops.
`† Placebo eye drops.
`Groups 1 and 2 were divided randomly by a controller (YY). Participants used AT for 2 weeks during the washout period, used oil eye drops or placebo
`for 2 weeks, and then switched eye drops for the next 2 weeks. Examinations were performed at the end of each 2-week period of treatment.
`
`eraged.24 Schirmer’s test was performed to measure tear secre-
`tion.18,25 Assessment of meibum expression and meibomian gland
`orifice obstruction was recorded as described above in the inclu-
`sion criteria section.7,9
`
`Study Design
`A randomized, prospective, double-blind, placebo-controlled
`crossover clinical study was performed after a 2-week wash-out
`period with preservative-free artificial tears (Soft Santear).26 Dur-
`ing the 2 periods of 2 weeks each, the patients were assigned
`randomly to receive oil eye drops or placebo eye drops six times
`daily (Table 2). We used normal saline solution as the placebo eye
`drops. At the end of each treatment period, i.e. the oil eye drop
`period and the placebo period the following examinations de-
`scribed below were performed.
`The examinations were carried out in the following order to
`avoid the influence of one procedure on another: subjective face
`scores,16 tear interference grading (DR-1,)22 tear evaporation
`test,23 fluorescein and rose bengal vital staining BUT measurement
`and assessment of meibomian gland orifice obstruction.7–9,16,24,25
`When the results of the oil eye drop period were better than those
`of the placebo period, the use of the oil eye drops was judged an
`improvement. Schirmer’s test was not performed during the study
`period to avoid any influence on the tear evaporation test, which is
`sensitive to ocular surface damage. Informed consent was obtained
`from all subjects. Institutional Review Board or Ethics Committee
`approval was not required for this study.
`
`Statistical Analysis
`All data are presented as means ⫾ standard deviation. Both eyes of
`each patient were studied separately. Symptoms and findings at the
`end of each treatment period, the oil eye drop period and the
`placebo period, were compared and analyzed by Wilcoxon’s
`signed rank test for nonparametric paired data. A P value of less
`than 0.05 was accepted as statistically significant.
`
`Results
`
`Figures 1 and 2 show results from representative case with tear
`interference images, tear break-up, and rose bengal staining at the
`end of the placebo period (Fig 1) and at the end of oil eye drop
`period (Fig 2). Each result shows improvement from the oil eye
`drop period compared with the placebo period.
`The results at the end of placebo period and oil eye drop period
`are shown in Table 3. Compared with the placebo period, the
`
`2032
`
`subjective face score improved during the oil eye drop period from
`6.7 ⫾ 1.6 to 5.5 ⫾ 1.8 (P ⫽ 0.004), along with the improvement
`of the other tear examinations. Tear interference grade decreased
`from 3.1 ⫾ 0.71 to 2.0 ⫾ 0.77 (P ⬍ 0.0001), tear evaporation rates
`during normal blinking decreased from 13 ⫾ 6.2 (10⫺7 g/second)
`to 11 ⫾ 7.5 (10⫺7 g/second; P ⫽ 0.01), and tear BUT was
`prolonged from 4.6 ⫾ 2.8 seconds to 12 ⫾ 3.5 seconds (P ⬍
`0.0001). The improvement of the subjective score also was ac-
`companied by improvements of the rose bengal staining score
`(from 2.2 ⫾ 0.85 to 1.4 ⫾ 1.3; P ⫽ 0.007) and the meibomian
`gland orifice obstruction score (from 2.2 ⫾ 0.38 to 1.6 ⫾ 0.63; P
`⫽ 0.002), but not accompanied by the improvement of the fluo-
`rescein staining score (from 0.4 ⫾ 0.81 to 0.13 ⫾ 0.33; P ⫽ 0.06).
`Results of these improvements between the MGD without
`aqueous tear deficiency group (n ⫽ 11) and the MGD with aqueous
`tear deficiency group (n ⫽ 9) showed no statistically significant
`difference. Administration of the oil eye drops was well tolerated, and
`none of the subjects reported irritation or severe blurring. None of the
`collected samples tested positive for bacteria or fungi. Blinding
`among participants, persons performing the intervention, outcome
`assessors, and data analyses were performed entirely by protocol.
`
`Discussion
`
`This study demonstrated a clear benefit of using low-con-
`centration homogenized oil eye drops for the treatment of
`MGD. Because MGD is a major cause of ocular irritation,
`symptomatic relief is an essential aspect of treatment. The
`improvement in symptoms was accompanied by improve-
`ment of objective findings, including tear interference im-
`age, tear evaporation, tear BUT, rose bengal staining, and
`meibomian gland orifice obstruction. The improvements
`were attributed to the effects of the oil eye drops because all
`patients used eye drops in a double-blind protocol. The eye
`drops were stable and well tolerated, and they did not cause
`irritation or blurred vision at room temperature. By using
`low-concentration oil, we succeeded in avoiding the usual
`complications of oil administration, such as blurred vision
`and a viscous sensation.
`To supplement the missing tear lipid in MGD, we used
`POE castor oil to solubilize the castor oil in distilled water
`in a low concentration.27 It has been reported that a hydro-
`philic lipid is required to enable oil to spread over the
`human tear aqueous layer and that polar phospholipids exist
`as a hydrophilic lipid in tears.2– 4,27 The POE castor oil is a
`derivative of castor oil, a polar lipid having a hydrophilic
`
`

`
`Goto et al
`
`䡠 Castor Oil Eye Drops for MGD
`
`Figure 1. Tear interference image, tear break-up, and rose bengal staining
`of the representative case at the end of the placebo period. A, The tear
`interference image shows a few colors and a nonuniform pattern. B, C,
`Slit-lamp photographs stained by fluorescein and rose bengal were taken
`immediately after eye opening. Note the remarkable tear break-up (B) and
`rose bengal staining of the ocular surface between the palpebral fissure (C).
`
`and a hydrophobic group, which is considered to be an ideal
`oil for artificial tears. Because the castor oil can spread over
`the ocular surface, it is expected to make the tear more
`stable, to decrease tear evaporation, and to decrease friction
`between the lid and ocular surface.28
`
`Figure 2. Tear interference image, tear break-up, and rose bengal staining
`of the same case as in Figure 1 at the end of oil eyedrop period. A, The tear
`interference shows a somewhat gray color and more uniform distribution.
`B, C, Slit-lamp photographs stained by fluorescein and rose bengal were
`taken immediately after eye opening. Tear break-up cannot be seen, and
`rose bengal staining was decreased. Figures 1B and 2B were enhanced
`digitally to show tear break-up using Adobe Photoshop (Adobe Systems
`Inc., San Jose, CA) from original photographs at brightness ⫺40 and
`contrast ⫹70. The reflex from the central cornea is the result of the digital
`enhancement.
`
`2033
`
`

`
`Ophthalmology Volume 109, Number 11, November 2002
`
`Table 3. Results of Crossover Study during Placebo and Oil Eye
`Drop Periods for the Treatment of Noninflamed Obstructive
`Meibomian Gland Dysfunction (Mean ⫾ Standard Deviation)
`
`Examination
`
`Face score (1–9)
`Tear interference
`grading (1–5)
`Tear evaporation
`rates (⫻10⫺7
`g/second)
`Fluorescein score
`(0–9)
`Rose bengal
`score (0–9)
`BUT (sec)
`Orifice
`obstruction
`(0–3)储
`
`Placebo
`Period*
`6.7 ⫾ 1.6
`3.1 ⫾ 0.71
`
`13 ⫾ 6.2
`
`Oil Eye
`Drop Period†
`5.5 ⫾ 1.8§
`2.0 ⫾ 0.77§
`
`P Value‡
`
`0.004
`⬍0.0001
`
`11 ⫾ 7.5§
`
`0.01
`
`0.40 ⫾ 0.81
`
`0.13 ⫾ 0.33
`
`2.2 ⫾ 0.85
`
`4.6 ⫾ 2.8
`2.2 ⫾ 0.38
`
`1.4 ⫾ 1.3§
`
`12 ⫾ 3.5§
`1.6 ⫾ 0.63§
`
`0.06
`
`0.007
`
`⬍0.0001
`0.002
`
`BUT tear break-up time.
`* Placebo eye drops.
`† Low-concentration homogenized castor oil eye drops.
`‡ Data were analyzed by Wilcoxon’s signed rank test for nonparametric
`paired data. P value was determined between placebo and oil eye drop
`period. A P value of ⬍0.05 was statistically significant.
`§ Statistically significant.
`储 Meibomian gland orifice obstruction score.
`
`Both castor oil and POE castor oil are commercially
`available, are inexpensive, and have been proven to have
`minimal topical toxicity for the eye.15,28
`It was surprising to us that only a minimal amount of
`castor oil changed the ocular surface abnormality a great
`deal. Moreover, there was evidence that tears over the
`ocular surface were stabilized based on the improved tear
`interference image and increase of tear BUT. The tear
`interference pattern significantly showed a more uniform
`pattern of spreading of the lipid film layer after administra-
`tion of the castor oil eye drops than with administration of
`the placebo.21,22 Even though the amount of castor oil was
`minimal, the oil eye drops were well homogenized in the
`tear lipid layer and later spread to form a more stable tear
`lipid layer and to improve, the ocular surface condition
`(Figs 1A and 2A).21,22,29
`There is a concern for the clinical application of castor
`oil eye drops because of their limited shelf life stability.17 It
`is expected that oil quality may change during storage room
`temperature and may be an obstacle for clinical application.
`When we stored the castor oil at 4°C for 4 weeks, the oil
`maintained stability and was well preserved. Moreover, it
`did not cause any irritation to the volunteers. However, after
`storage of the castor oil eye drops at 60°C for 4 weeks, the
`ph changed to 5.8 and caused slight irritation to the patients.
`Castor oil itself is a very effective treatment for ocular
`surface diseases, but the storage method should be consid-
`ered before broad clinical application because of the insta-
`bility of the oil at higher temperatures.
`In conclusion, we have documented a beneficial effect of
`castor oil for the treatment of ocular surface disorders with
`MGD. Although the concept of administering oily eye drops
`has not been considered seriously, the application of castor
`
`2034
`
`oil eye drops should be considered more for the ideal
`artificial tears. Theoretically, patients with aqueous tear
`deficiency can benefit from the application of these artificial
`tears because MGD needs to be supplemented by oil or
`some other ingredient that can stabilize the ocular surface,
`decrease tear evaporation, and decrease friction between the
`lid and ocular surface. Based on study, it is necessary to
`develop a means of maintaining the stability of artificial
`tears with castor oil in rooms with higher temperatures;
`however, this randomized double-blind study clearly shows
`the benefit of the addition of castor oil to artificial tears.
`
`References
`
`1. Mishima S, Maurice DM. The oily layer of the tear film and
`evaporation from the corneal surface. Exp Eye Res 1961;1:
`39 – 45.
`2. Holly FJ. Tear film physiology. Am J Optom Physiol Opt
`1980;57:252–7.
`3. Tiffany JM. The role of meibomian secretion in the tears.
`Trans Ophthalmol Soc U K 1985;104:396 – 401.
`4. Tiffany JM. The lipid secretion of the meibomian glands. Adv
`Lipid Res 1987;22:1– 62.
`5. Driver PJ, Lemp MA. Meibomian gland dysfunction. Surv
`Ophthalmol 1996;40:343– 67.
`6. Lemp MA. Report of the National Eye Institute/Industry
`Workshop on Clinical Trials in Dry Eyes. CLAO J 1995;21:
`221–32.
`7. Shimazaki J, Sakata M, Tsubota K. Ocular surface changes
`and discomfort in patients with meibomian gland dysfunction.
`Arch Ophthalmol 1995;113:1266 –70.
`8. Lee SH, Tseng SCG. Rose bengal staining and cytologic
`characteristics associated with lipid tear deficiency. Am J
`Ophthalmol 1997;124:736 –50.
`9. Shimazaki J, Goto E, Ono M, et al. Meibomian gland dys-
`function in patients with Sjo¨gren syndrome. Ophthalmology
`1998;105:1485– 8.
`10. Key JE. A comparative study of eyelid cleaning regimens in
`chronic blepharitis. CLAO J 1996;22:209 –12.
`11. Korb DR, Greiner JV. Increase in tear film lipid layer thick-
`ness following treatment of meibomian gland dysfunction. In:
`Sullivan DA, ed. Lacrimal Gland, Tear Film, and Dry Eye
`Syndromes. New York: Plenum Press, 1994;293– 8.
`12. Smith RE, Flowers CW Jr. Chronic blepharitis: a review.
`CLAO J 1995;21:200 –7.
`13. Rieger G. Lipid-containing eye drops: a step closer to natural
`tears. Ophthalmologica 1990;201:206 –12.
`14. Tiffany JM. Lipid-containing eye drops [letter]. Ophthalmo-
`logica 1991;203:47–9.
`15. Japanese Pharmaceutical Excipients 1993. Tokyo: Yakuji
`Nippo, 1994; 109.
`16. Tsubota K, Goto E, Fujita H, et al. Treatment of dry eye by
`autologous serum application in Sjo¨gren’s syndrome. Br J
`Ophthalmol 1999;83:390 –5.
`17. Awata T. Stability of The Eye Drops. In: Ohashi Y, ed. The
`Eye Drops. Tokyo: Medical View Co, 1996;15–23.
`18. Xu KP, Yagi Y, Toda I, Tsubota K. Tear function index. A
`new measure of dry eye. Arch Ophthalmol 1995;113:84 – 8.
`19. Fox RI, Robinson CA, Curd JG, et al. Sjo¨gren’s syndrome.
`Proposed criteria for classification. Arthritis Rheum 1986;29:
`577– 85.
`20. Robin JB, Jester JV, Nobe J, et al. In vivo transillumination
`biomicroscopy and photography of meibomian gland dysfunc-
`tion. A clinical study. Ophthalmology 1985;92:1423– 6.
`
`

`
`Goto et al
`
`䡠 Castor Oil Eye Drops for MGD
`
`21. Yokoi N, Takehisa Y, Kinoshita S. Correlation of tear lipid
`layer interference patterns with the diagnosis and severity of
`dry eye. Am J Ophthalmol 1996;122:818 –24.
`22. Yokoi N, Mossa F, Tiffany JM, Bron AJ. Assessment of
`meibomian gland function in dry eye using meibometry. Arch
`Ophthalmol 1999;117:723–9.
`23. Tsubota K, Yamada M. Tear evaporation from the ocular
`surface. Invest Ophthalmol Vis Sci 1992;33:2942–50.
`24. Toda I, Tsubota K. Practical double vital staining for ocular
`surface evaluation [letter]. Cornea 1993;12:366 –7.
`25. van Bijsterveld OP. Diagnostic tests in the sicca syndrome.
`Arch Ophthalmol 1969;82:10 – 4.
`
`26. Rulo AH, Greve EL, Geijssen HC, Hoyng PFJ. Reduction of
`intraocular pressure with treatment of latanoprost once daily in
`patients with normal-pressure glaucoma. Ophthalmology
`1996;103:1276 – 82.
`27. Mitsui T. Physiochemistry of the Cosmetics. 1st ed. Tokyo:
`Nanzando, 1993; 128 –56.
`28. Nagai K. Polyoxyethylene castor oil derivatives. In: Saito
`Y, ed. Handbook of Pharmaceutical Excipients, Japanese
`Edition. Tokyo: The Pharmaceutical Society of Japan,
`1986;345– 8.
`29. Korb DR, Baron DF, Herman JP, et al. Tear film lipid layer
`thickness as a function of blinking. Cornea 1994;13:354 –9.
`
`2035